Resilient and Efficient Aerial Wireless Sensor Networks Using Pinching-Antenna Systems

I. Introduction

• Pinching antennas provide spatially adjustable radiation points intended to focus signals toward users and improve link quality and blockage robustness, motivating their study for power-limited aerial sensor networks.The introduction also notes their potential to establish strong line-of-sight links and reduce free-space path loss with UAVs.Source: §I · pp. 1–2
• The Fiedler value measures network connectivity: a nonzero value indicates every sensor can reach at least one UAV, while a higher value implies greater resilience.It is the second-smallest eigenvalue of the graph Laplacian.Source: §I · p. 2
• The paper formulates pinching-antenna placement to maximize the network Fiedler value, subject to inter-element spacing constraints.A heuristic algorithm places elements in the uplink and downlink waveguides to pursue this objective.Source: §I · pp. 2–3
• The UAV-deployment problem minimizes the number of deployed UAVs while enforcing a minimum Fiedler value and practical placement and antenna-spacing constraints.An efficient algorithm determines the minimum required UAV count and their deployment locations.Source: §I · pp. 2–3

II. System Model

• The proposed network uses a base station with M uplink and M̄ downlink pinched dielectric waveguides to serve S sensors and U UAVs.Direct sensor–UAV communication forms an initial graph with S + U vertices.Source: §II · pp. 3–4, Figure 1 · p. 3
• Pinching-element positions on each uplink and downlink waveguide are decision variables that determine additional communication edges through the base station.The resulting graph retains the initial vertices and combines direct edges with the added base-station edges.Source: §II · p. 4

A. Communication Channels Model

• The uplink and downlink waveguide radiation vectors are represented by antenna phase shifts, with each phase shift associated with a pinching element.The uplink phases are θk,m and the downlink phases are ¯θk, ¯m.Source: A. Communication Channels Model · p. 4
• For uplink, the channel from each sensor to a waveguide’s pinching elements is collected in a vector, and the effective channel concatenates the links across uplink waveguides.The model identifies ζ with a coefficient proportional to pinching-antenna effective surface, αi with shadowing, κ with wave number, and δi,k(xm) with sensor-to-element distance.Source: A. Communication Channels Model · p. 4
• For downlink, the channel from each waveguide’s pinching elements to a UAV is collected in a vector, and the effective channel combines links across downlink waveguides.The element-to-UAV distance is denoted ¯δk,j(¯x ¯m).Source: A. Communication Channels Model · p. 5
• Uplink and downlink transmissions use orthogonal frequency- or time-domain channels, and their SNRs are defined using the respective transmit powers and additive-noise powers.The uplink SNR is evaluated at the base station for each sensor, while the downlink SNR is evaluated at each UAV.Source: A. Communication Channels Model · p. 5
• A sensor–UAV graph edge is added only when the sensor’s uplink SNR meets γu and the UAV’s downlink SNR meets γd.The thresholds represent the minimum SNRs for reliable communication with the base station and a UAV, respectively.Source: A. Communication Channels Model · p. 5

• The graph Laplacian is defined as L ≜ D − A, where D is the degree matrix and A is the adjacency matrix.Source: B. Network Connectivity Metric · p. 5
• The Laplacian is real, symmetric, and positive semi-definite, with smallest eigenvalue L1 equal to zero and corresponding to the all-ones eigenvector.Source: B. Network Connectivity Metric · p. 6
• The second-smallest eigenvalue L2, or Fiedler value, is the graph’s algebraic connectivity; it is nonzero when the graph is connected and increases monotonically with the edge set [9],.Source: B. Network Connectivity Metric · p. 6

• The section frames the design problem as developing a pinching-antenna framework for resilient and efficient aerial wireless sensor networks.Source: §III · p. 6

A. Resilient Aerial Wireless Sensor Networks

• The placement objective is to maximize the Fiedler value of the network graph by optimizing uplink and downlink pinching-antenna locations.This objective is stated as a means to maximize connectivity and maintain a resilient network.Source: A. Resilient Aerial Wireless Sensor… · p. 6
• Neighboring pinching elements on the uplink and downlink waveguides must meet minimum spacing constraints to prevent mutual coupling.Constraints (9b) and (9c) impose the minimum separations.Source: A. Resilient Aerial Wireless Sensor… · p. 6
• Pinching elements must be placed within the length of their corresponding waveguides.Constraints (9d) and (9e) enforce this placement bound; (9e) explicitly bounds downlink positions between 0 and the waveguide length.Source: A. Resilient Aerial Wireless Sensor… · p. 6

B. Efficient Aerial Wireless Sensor Networks

• The objective is a connected aerial sensor network that maintains the minimum connectivity threshold with the fewest deployed UAVs.Source: B. Efficient Aerial Wireless Sensor… · p. 6
• The formulation optimizes UAVs’ 3D placements subject to a connectivity threshold ς > 0 and an allowable altitude limit.Source: B. Efficient Aerial Wireless Sensor… · p. 7

IV. Solution Approach

• Waveguides are divided into equal segments whose lengths meet minimum inter-element spacing, with each pinching element placed at a segment center.This discretization satisfies constraints (9b)–(9e).Source: §IV · p. 7
• The placement heuristic selects the lowest-degree sensor or UAV and assigns a pinching element to the nearest unoccupied segment on the corresponding uplink or downlink waveguide.It excludes a waveguide from further selection once all its elements have been placed.Source: §IV · pp. 7–8
• The UAV deployment algorithm starts with U = 1, clusters sensors into U groups, places each UAV at its cluster center, and increases U until constraint (10b) is satisfied.The UAV altitude is obtained with constraint (10c) in mind, using the farthest sensor’s distance from its cluster center and constant θopt.Source: §IV · p. 8

A. Computational Complexity

• Segmenting the waveguides takes O(N + N_bar) operations; sensor and UAV selection take O(S), O(N), O(U), and O(N_bar), respectively.Updating the edge set takes O(SK + U K_bar) operations.Source: A. Computational Complexity · p. 8
• Algorithm 2 solves P2 in O(U K_tilde[SK + U K_bar + N + N_bar] + U^2SQ) operations, where K_tilde = max{K, K_bar}.Its sensor clustering and UAV-location steps each require O(U^2SQ) operations.Source: A. Computational Complexity · p. 8
• Algorithm 1 solves P1 in O(K_tilde[SK + U K_bar + N + N_bar]) operations, where K_tilde = max{K, K_bar}.The operation count includes waveguide segmentation, degree-based sensor and UAV selection, and edge-set updates.Source: A. Computational Complexity · p. 8

• The simulations randomly distribute sensors across a 120 × 20 m area, with uplink and downlink waveguides alternatively equally spaced along the y-axis.Source: §V · p. 10
• The proposed solution achieves better connectivity than equally spaced placement and conventional antennas, approaches exhaustive search, and benefits from deploying more UAVs.Source: Figure 2 · p. 11, §V · p. 11
• Algorithm 2 requires slightly more UAVs as sensor count grows and more UAVs as threshold ς increases, while more waveguides and pinching elements reduce the required count.Source: §V · pp. 11–12

VI. Conclusion

• Simulations showed that the pinching-antenna system improves network connectivity and efficiency compared with the conventional system.Efficiency is measured by the number of deployed UAVs required to connect the network.Source: §VI · p. 12, Figure 3 · p. 12
• The heuristic pinching-antenna placement algorithm maximizes network connectivity by optimizing element placement in the uplink and downlink waveguides.The Fiedler value is used as the network resilience indicator.Source: §VI · p. 12
• The iterative UAV deployment algorithm finds the minimum UAV count, UAV locations, and pinching-antenna placement needed to connect the network.Source: §VI · p. 12